Children with neuropsychiatric conditions, such as autism spectrum disorders (ASD), are at high risk for insomnia, and pharmacological treatments for insomnia in this population are acutely needed. Lack of restorative sleep may exacerbate disabling behaviors associated with ASD. Treatments that promote sleep in children with ASD may ameliorate these behaviors as well as strengthen the ability of parents to cope with this disorder and to assist their children in overcoming their disabilities. Our long-term objectives are to determine if supplemental melatonin is an effective treatment for insomnia in children with ASD and whether the use of melatonin to treat insomnia improves these children's daytime behaviors and reduces levels of parental stress. Although generally accepted as safe and well tolerated, melatonin's efficacy in ASD has not been proven in controlled studies, and treatment strategies related to dosage vary widely. Prior to carrying out a Phase III Randomized Clinical Trial (RCT) that will test these objectives, critical design issues need to be addressed. This proposed Preliminary Clinical Trials Grant will allow us to design an effective RCT. The Specific Aims of this Preliminary Clinical Trials Grant are to: (1) Optimize the intervention strategy for administering supplemental melatonin in relation to safety, tolerability, and effect on sleep latency in children with ASD;(2) Characterize the pharmacokinetic profile of supplemental melatonin;and (3) Pilot a group of behavioral and parental stress scales in preparation for an RCT. To achieve these aims, we will treat 30 children with ASD, ages 4-10, with supplemental melatonin (1-9 mg) using a 16-week optional titration design. This design allows parents to stop titration at a dose in which their children respond to treatment and at which no significant adverse events occur. A subset of 12 children will undergo pharmacokinetic studies. We will determine the response of melatonin dose on sleep latency using polysomnography and actigraphy, a validated and non-intrusive measure of sleep onset and duration. Adverse events will also be captured.